By: Leor Sapir, Mungeri Lal
Published: Sep 19, 2024
The Department of Health and Human Services’ documented failures to hold gender medicine to scientific standards have happened under both Republican and Democratic administrations.
In 2015, the Centers for Medicare & Medicaid Services (CMS) accepted a formal request to initiate a national coverage analysis for “gender-reassignment surgery.” When making these coverage determinations, CMS is legally obligated to evaluate relevant clinical evidence and answer the question: Is the evidence sufficient to conclude that the application of the item or service under study will improve health outcomes for Medicare patients? In June 2016, CMS released its 109-page analysis, which it made open for public comment. The agency published its final decision memo two months later. The differences between the two documents were revealing—and disturbing.
At first sight, the summaries of both memos seemed similar. Each mentioned that the CMS was not issuing a national coverage determination on “gender-reassignment surgery” for Medicare beneficiaries with gender dysphoria. Such determinations, the CMS said, would continue to be made by local contractors on a case-by-case basis. On closer examination, however, the final document included substantial changes. These were not corrections. They amounted to a systematic effort to scrub any reference to the evidence of the harms associated with these surgeries.
The agency’s shifting analysis of gender surgery is glaring enough. But it is just one of several examples, between 2016 and the present day, of how key figures and agencies within the U.S. Department of Health and Human Services have misled the American public about the evidence for “gender-affirming care.”
Start with the tale of two memos. To conduct a proper analysis of the clinical evidence, CMS identified a large number of publications related to “gender-reassignment surgery.” As CMS explained:
Studies with robust study designs and larger, defined patient populations assessed with objective endpoints or validated test instruments were given greater weight than small, pilot studies. Reduced consideration was given to studies that were underpowered for the assessment of differences or changes known to be clinically important. Studies with fewer than 30 patients were reviewed and delineated, but excluded from the major analytic framework. Oral presentations, unpublished white papers, and case reports were excluded. Publications in languages other than English were excluded.
Thirty-three publications, ranging from 1979 to 2015, were eventually identified and included. Twenty-four of these studies were conducted in Europe, compared with just six from the United States—not surprising, considering that European nations have historically kept better track of such medical interventions. Patients in the studies were typically in their twenties and thirties. CMS also consulted the guidelines and recommendations of medical groups, including the American Psychological Association, the Endocrine Society, and the U.S. Institute of Medicine. The agency considered evidence from any listed clinical trials at the time and solicited public comments.
All of this is to say that CMS’s analysis was rigorous. It went through the 33 studies, explaining their methodology, results, strengths, and weaknesses. The analyses were more detailed than those found in some of the systematic reviews carried out in the medical literature, which have space limitations. The individual studies varied widely in their rigor and reporting.
Though the results of these studies were somewhat inconsistent, CMS found that the best studies all pointed to negative outcomes.
The proposed analysis—the first memo that CMS released—identified one particular study from Sweden as “the most comprehensive study with functional endpoints of the 33 studies reviewed” and devoted considerable space to discussing it. What differentiated the Swedish study from the others? Unlike most other studies that relied on clinic samples (a convenient choice for researchers but one that introduces the risk of bias) for a limited amount of time (thereby missing long-term consequences), the Swedish study tracked all patients who had undergone “gender-reassignment surgery” over a 30-year interval and compared them with 6,480 matched controls from the general Swedish population.
The proposed memo explained why it considered this particular study comprehensive. The data came from a compulsory national database. The study followed all patients nationally across all clinics for over 30 years. It used the same, consistent criteria for everyone permitted to undergo surgery, allowing an accurate comparison across patients. It captured objective outcomes, namely mortality and regret (narrowly defined as a recorded request for surgical reversal), as endpoints. And it included a full record of all suicide and suicide attempts in the clinical data in the Swedish National Patient Register.
The Swedish study presented an alarming picture of life after “gender-reassignment surgery.” To appreciate why, consider two additional features. First, the study used “highly vetted patients”—only those who passed a battery of physiological and psychological tests were eligible to undergo the surgery. Second, surgeries were conducted only “at select centers with integrated care . . . in which there is sequential evaluation of patients for progressively more invasive interventions.” Thus, individuals who received these surgeries seemed comparatively better adjusted at baseline and were well positioned to enjoy all-around care following their procedures.
Yet even after receiving “gender-reassignment surgery,” transsexuals in the Swedish study were 19.1 times more likely to die by suicide than were members of the control group. The authors adjusted for psychiatric history, among other things. According to CMS, the study could not isolate the impact of “reassignment surgery,” but it did find that “[r]eassignment surgery does not return patients to a normal level of morbidity risk and that the morbidity risk is significant even in highly vetted patients in a structured care system.” In addition, the divergence in survival rates from the rest of the Swedish population did not become apparent until at least ten years, an interval much longer than the follow-up period of other studies. The survival rates at the 20-year follow-up were: female controls 97 percent, male controls 94 percent, “female-to-male” patients 88 percent, and “male-to-female” patients 82 percent.
The ultimate CMS report systematically downplayed these findings.
CMS’s two memos differed in multiple ways. First, and less suspiciously, the final memo contained a large section noting and responding to public comments. This was to be expected, though the comments were interesting. One group of commenters “suggested that CMS should recommend the WPATH Standards of Care (WPATH) as the controlling guideline for” gender surgery. Also in the final memo, in a section titled “Evidence Summary and Analysis,” CMS notes that several commenters “disagreed with our summary of the clinical evidence and analysis” in the NCA of the proposed memo and that some of them felt that “the overall tone of the review was not neutral and seemed biased or flawed.” The CMS team disagreed with this contention, insisting that “the summary and analysis of the clinical evidence are objective” and that, “As with previous [national coverage analyses], our review of the evidence was rigorous and methodical.”
Despite these pronouncements, however, CMS made substantial changes to the final memo regarding the analysis of clinical evidence. Both the proposed and final memos include a section called “Knowledge Gaps” that discusses the limitations of the current evidence in the literature.
This section was much larger in the first memo. And it included a crucial observation: “The results were inconsistent, but negative in the best studies, i.e., those that reduced confounding by testing patients prior to and after surgery and which used psychometric tests with some established validation in other large populations” (emphasis added). That entire sentence was struck from the final memo.
Also gone were large swathes of text that covered the Swedish paper, described in the proposed memo as “the most comprehensive study with functional endpoints of the 33 studies reviewed.” The final memo never explained why the Swedish study had been considered so comprehensive.
Similarly missing was the entire section on “Mortality and Regret as Endpoints.” In the proposed memo, that section had argued that death and regret (measured by surgical reversal) represented more “objective” measuring sticks than did psychometric measures. Gone, too, were statements from the Swedish study that “[r]eassignment surgery does not return patients to a normal level of morbidity risk,” and that this risk “is significant even in highly vetted patients in a structured care system.”
The final memo made subtler changes, too. A sentence that appeared in the proposed memo reappears in the final memo, but in modified form. “Much of the available research has been conducted in highly vetted patients at select care programs integrating psychotherapy, endocrinology, and various surgical disciplines and operating under European medical management and regulatory structures.” The struck reference to Europe and its regulatory structures—or even the seemingly innocuous fact that most of the studies emanate from Europe—apparently had to go. Readers familiar with the general European trend of restricting “gender-affirming care” can draw their own inferences.
Gone, finally, from the final memo is a discussion under patient care about the necessary surgical expertise and care settings for the administration of these procedures. “The surgical expertise and care setting(s) required to improve health outcomes in adults with gender dysphoria remain(s) uncertain. The selection of a particular surgeon could become an important variable if subjective outcomes depend on functional surgical results. . . . Many of these procedures involve complicated gynecologic, urologic surgical techniques accompanied by significant risk. . . . Most of the studies for reassignment surgery have been conducted in northern Europe at select centers with integrated care (psychological, psychiatric, endocrinologic, and surgical) in which there is sequential evaluation of patients for progressively more invasive interventions.” (Emphasis added.) This omission likely occurred because the recommendation amounted, de facto, to a barrier to access—at the time of writing, very few surgeons were qualified to carry out these procedures in North America—and observed that such procedures are highly complex and risky.
It’s hard to avoid the conclusion that, in its final memo, CMS tried its best to paint a rosy picture of “gender-reassignment surgery.” It did so by misleading the public—not only mischaracterizing the state of the evidence but also, it seems, leaving out details about what these procedures actually involve and what risks they pose.
It wasn’t the last time that HHS would provide cover for dubious conduct in the gender-medicine field. On April 5, 2019, a group of clinicians wrote a letter to Jerry Menikoff, then the director of the Office for Human Research Protections at HHS, alerting him to ethical concerns regarding an ongoing NIH-funded study on the use of “gender-affirming” cross-sex hormones in children as young as age eight. The study’s principal investigator, Johanna Olson-Kennedy, is one of the most prominent figures in the world of youth gender medicine and is known for (among other things) devising dubious research to cement the practice of double mastectomy in teenage girls.
The authors of the letter to Menikoff emphasized several critical problems in Olson-Kennedy’s study, including the lack of a control group, the extremely young age of eligibility (much lower than 16, which is what European countries were recommending at the time), the dangerously high doses participants were being given, the risks of long-term harms including infertility, and the lack of informed consent.
The clinicians did not hear back from Menikoff, but they did receive a response from Diana Bianchi, director of NIH’s National Institute of Child Health and Human Development, who dismissed their concerns. According to Bianchi, the participants in the study “sought and obtained the hormonal therapies independent of the [study’s] protocol. Therefore, termination of the protocol would not end the treatments; rather it would only end the compilation of data needed to advance scientific understanding of the risks and likely outcomes of those treatments.” In other words, according to NIH, there was no ethical problem with Olson-Kennedy and her colleagues conducting an uncontrolled experiment on vulnerable young teens because they were not technically conducting an experiment; they were merely observing the effects of an already proceeding medical intervention.
Bianchi’s assumption that the researchers intended to “compil[e]. . . data needed to advance scientific understanding of the risks and likely outcomes of those treatments” turned out to be overly optimistic. When the researchers—who thus far have received over $9 million for this study from the taxpayer-funded NIH—published the first round of their findings in the New England Journal of Medicine in 2023, their study contained serious methodological problems. The results were far from impressive and mischaracterized by the authors, who also seemed unconcerned that two of their 315 adolescent participants had committed suicide after commencing hormones.
In July 2020, the Agency for Healthcare Research and Quality (AHRQ), an entity within HHS responsible for conducting systematic reviews of evidence, received a request for just such a review from the American Academy of Family Physicians (AAFP) on “Treatments for Gender Dysphoria in Transgender Youth.” Unfortunately, the agency wound up shirking its duty to provide a major medical association with accurate information about this dubious practice.
To ensure “gender-affirming care” was not already being studied systematically, AHRQ searched a database in which researchers register their protocols. On doing so, it apparently learned that researchers at Johns Hopkins University were already conducting several reviews on behalf of WPATH. (Court documents would eventually reveal that WPATH manipulated those evidence reviews.) In any case, WPATH had commissioned the reviews from Hopkins as part of a process to revise its “Standards of Care,” a document that has received wide deference from doctors, clinics, insurance carriers, state and federal health authorities, lawyers, and journalists.
Shortly after receiving the request from the association of family-medicine doctors, AHRQ reached out to the Hopkins team to inquire about whether they were examining outcomes for minors. The lead Hopkins researcher wrote back that they had “found little to no evidence” about the use of medical gender transition in “children and adolescents,” but that they were “having issues” with their “sponsor [WPATH] trying to restrict our ability to publish.”
Six months later, on January 8, 2021, AHRQ released a memo explaining why it would not conduct a systematic review of youth gender transition. “[W]e found protocols for two systematic reviews that address[] portions of the [AAFP] nomination,” the agency explained, “and an insufficient number of primary studies exist to address the remainder of the nomination.” One of the two protocols the agency cited was the WPATH review itself. The second “protocol” AHRQ cited was not a systematic review protocol at all but a single study on double mastectomy performed on “transmasculine youth” at “one pediatric center.” The study was published in the International Journal of Transgenderism, the official journal of WPATH.
In other words, a branch of HHS dedicated to evaluating empirical evidence told a major medical association of family practitioners that a review was underway—despite having learned six months earlier that the sponsor of that review was actively trying to prevent its publication.
That AHRQ listed these WPATH-affiliated “protocols” as a reason not to conduct a systematic review is suspicious. It is especially so considering that, in the decision memo, AHRQ noted, “There is a lack of current evidence-based guidance for the care of children and adolescents who identify as transgender, particularly regarding the benefits and harms of pubertal suppression, medical affirmation with hormone therapy, and surgical affirmation. While there are some existing guidelines and standards of care, most are derived from expert opinion or have not been updated recently.” Why, then, credit WPATH’s attempts to make the record show otherwise?
The agency’s decision also seemed to be at odds with its own criteria for when to do a systematic review. In an appendix to its decision, the agency explained that a systematic review is appropriate when, among other things, there is no “recent high-quality systematic review . . . on this topic” and when “clinical benefits and potential harms” from some intervention are in question. The family-practitioner association mentioned some of those harms—including cancer and cardiovascular disease—in its letter; in its exchange with Hopkins, AHRQ had asked if the university’s team was planning to look at these risks. “I don’t think any of the planned manuscripts would be an overlap [with the AAFP questions],” the Hopkins lead author replied.
Months after the exchange between AHRQ and the Hopkins team, Rachel Levine—the Biden administration’s transgender assistant secretary for health at HHS, who advocates “gender-affirming” interventions in minors—pressured WPATH to eliminate age minimums for medical procedures, including surgeries, in its soon-to-be-released updated standards of care (known as SOC-8). WPATH complied, issuing a correction several days after publication and doing away with age minimums. After news of Levine’s role in shaping SOC-8 broke, the Biden White House half-heartedly walked back its support for gender surgeries for kids. Levine, who is a pediatrician, is currently scheduled to be a keynote speaker in the American Academy of Pediatrics’ upcoming national conference in Orlando, Florida.
Around the same time that Levine was meddling in WPATH’s guideline development process, the Food and Drug Administration, itself an HHS agency, released a warning about brain tumor–like symptoms that can result in vision loss associated with use of puberty blockers. The conservative group America First Legal promptly served the FDA with a Freedom of Information Act request for FDA communications regarding the off-label use of puberty blockers in transgender procedures.
The request returned correspondence between Reuters journalists, who were working on a series of articles about youth gender medicine, and FDA officials. “I saw mention of a 2017 FDA safety review of these drugs related to CPP [central precocious puberty],” the journalist wrote. “There are questions about bone health, brain development and fertility. Did anything come out of that? Has FDA done anything in relation to their off-label use for gender dysphoria in children?”
The official who received the email from Reuters forwarded it to colleagues for help. An official whose email signature designates her as “Clinical Team Leader” for the “Division of General Endocrinology” wrote back to her colleague:
Most of these patients had CPP but a handful were transgender kids using the drugs off label. We found no effect on bone (after factoring in catch-up growth), including no increase in fracture risk. We did find increased risk of depression and suicidality, as well as increased seizure risk and we issued [safety labeling changes] to the entire class for these [Adverse Events] (added to [Warnings and Precautions] in 2017). . . . Regarding use of GnRH agonists in the transgender population, no company [that produces puberty blockers] has come in for this indication to date. DUOG [Division of Urology, Obstetrics, and Gynecology] has done a patient listening session with trans kids and separately with trans adults, which I participated in, and there is definitely a need for these drugs to be approved for gender transition, as they are typically not covered by insurance and are expensive out of pocket. It was my understanding that DUOG would take these applications if and when any do come in.
These four examples (and there are likely more) tell a grim story. Key figures and departments within HHS have repeatedly failed to live up to the agency’s mission to improve “the health, safety, and well-being of America.” These failures may have been intentional. They may have been the result of incompetence. They may have reflected unwarranted trust in researchers presumed to be concerned with scientific evidence above political agendas. Or they may reflect standard bureaucratic pathologies, from excessive red tape to deficiencies in intra-agency communication and coordination. Whatever the reason, the outcome is clear: HHS has misled the public about gender medicine under Democratic and Republican administrations alike.